U.S. patent number 5,461,682 [Application Number 08/048,275] was granted by the patent office on 1995-10-24 for image filing apparatus providing image data suitable for several input/output devices.
This patent grant is currently assigned to Ricoh Company Ltd.. Invention is credited to Keiichi Nomura.
United States Patent |
5,461,682 |
Nomura |
October 24, 1995 |
Image filing apparatus providing image data suitable for several
input/output devices
Abstract
An image filing apparatus according to the invention comprises a
unit for inputting an image, a plurality of data compressing and
extending units each for compressing the input image data to store
the compressed image data and each for extending the compressed
image data to output the extended image data to at least one of
image outputting units such as an image display, an image printer
and the like, and a control unit for selecting one of the plurality
of data compressing and extending units in accordance with an image
type of the input image to supply it to the selected data
compressing and extending unit upon the compression and for
supplying the compressed image data to the selected data
compressing and extending unit upon the extension. Thus, the
optimum compressing and extending method is selected and applied to
the input image data in accordance with the type of the image
outputting unit from which the image data is to be output.
Inventors: |
Nomura; Keiichi (Kawasaki,
JP) |
Assignee: |
Ricoh Company Ltd. (Tokyo,
JP)
|
Family
ID: |
14976636 |
Appl.
No.: |
08/048,275 |
Filed: |
April 14, 1993 |
Foreign Application Priority Data
|
|
|
|
|
Apr 20, 1992 [JP] |
|
|
4-128110 |
|
Current U.S.
Class: |
382/232; 358/403;
358/426.02 |
Current CPC
Class: |
G06T
9/007 (20130101); H04N 1/32502 (20130101); H04N
1/32529 (20130101); H04N 1/40062 (20130101); H04N
1/41 (20130101); H04N 2201/0081 (20130101); H04N
2201/0082 (20130101); H04N 2201/0089 (20130101) |
Current International
Class: |
G06T
9/00 (20060101); H04N 1/32 (20060101); H04N
1/41 (20060101); H04N 1/40 (20060101); G06K
009/36 (); H04N 001/41 () |
Field of
Search: |
;382/56
;358/426,430,403,432,433,451 ;345/202 ;395/114,934 |
References Cited
[Referenced By]
U.S. Patent Documents
Primary Examiner: Moore; David K.
Assistant Examiner: Johns; Andrew W.
Attorney, Agent or Firm: Cooper & Dunham
Claims
What is claimed is:
1. An image filing apparatus comprising:
image reading means for reading an image;
a plurality of image data compressing and extending units, each of
said plurality of image data compressing and extending units
respectively corresponding to one of a plurality of image
outputting units and being adapted to compress and extend image
data read by said image reading means by a data compressing and
extending method different from each other; and
control means for selecting an image data compressing and extending
unit from said plurality of image data compressing and extending
units with respect to one of said plurality of image outputting
units in accordance with an output form of said one of said
plurality of image outputting units, and outputting the image data
extended by the selected image data compressing and extending unit
to said one of said plurality of image outputting units,
wherein said apparatus further comprises images processing means
for changing the scale of the image read by the image reading means
in accordance with the number of dots in an image displaying unit,
the image processing means changing the scale of the image before
compression of the image data.
2. An image filing apparatus according to claim 1, wherein said
control means comprises determining means for determining an image
type of the image read by the image reading means using an array of
DCT coefficients.
3. An image filing apparatus according to claim 2, wherein said
determining means computes an N.times.N array of mean magnitudes of
DCT coefficients in a plurality of N.times.N arrays of a target
area and reference areas located adjacent to the target area of the
image and thereby determining an image type of the target area of
the image from a distribution of the mean magnitudes of elements in
the computed N.times.N array.
4. An image filing apparatus according to claim 3, wherein when
said distribution has a peak in a predetermined region of the
array, said determining means determines that the target area of
the image includes meshes.
5. An image filing apparatus according to claim 4, wherein when
said distribution has a peak at a predetermined diagonal element of
the array, said determining means conclude that the target area of
the image includes meshes.
6. An image filing apparatus according to claim 3, wherein said
control means selects the optimum image data compressing and
extending unit on the basis of determining result of said
determining means upon the compression of the image data.
7. An image filing apparatus according to claim 2, wherein said
control means selects the image data compressing and extending unit
on the basis of a determining result of said determining means upon
the compression of the image data.
8. A method for filing images comprising the steps of:
reading an image;
providing a plurality of image data compressing and extending
units, each of said plurality of image data compressing and
extending units respectively corresponding to one of a plurality of
image outputting units and being adapted to compress and extend
image data by a data compressing and extending method different
from each other;
selecting an image data compressing and extending unit from said
plurality of image data compressing and extending units with
respect to one of said plurality of image outputting units in
accordance with an output form of said one of said plurality of
image outputting units;
compressing and extending the image read in said reading step using
the selected image data compressing and extending unit; and
outputting the image data compressed and extended by the selected
image data compressing and extending unit to said one of said
plurality of image outputting units,
wherein said method further comprises the step of changing the
scale of the image data read in the reading step in accordance with
the number of dots in an image displaying unit prior to said step
of compressing and extending.
9. A method for filing images according to claim 8, wherein said
step of selecting comprises the step of determining an image type
of the image read in the reading step using an array of DCT
coefficients.
10. A method for filing images according to claim 9, wherein said
step of determining comprises computing an N.times.N array of mean
magnitudes of DCT coefficients in a plurality of N.times.N arrays
of a target area and reference areas located adjacent to the target
area of the image and thereby determining an image type of the
target area of the image from a distribution of the mean magnitudes
of elements in the computed N.times.N array.
11. A method for filing images according to claim 10, wherein when
said distribution has a peak in a predetermined region of the
array, said determining step determines that the target area of the
image includes meshes.
12. A method for filing images according to claim 11, wherein when
said distribution has a peak at a predetermined diagonal element of
the array, said determining step concludes that the target area of
the image includes meshes.
13. A method for filing images according to claim 10, wherein said
step of selecting an image data compressing and extending unit from
said plurality of image data compressing and extending units is
done on the basis of said step of determining upon the compression
of the image data.
14. A method for filing images according to claim 9, wherein said
step of selecting an image data compressing and extending unit from
said plurality of image data compressing and extending units is
done on the basis of said step of determining upon the compression
of the image data.
Description
BACKGROUND OF THE INVENTION
1. Field of the Invention
The present invention relates to the improvement of an image filing
apparatus such as an optical disk, or an electronic filing
apparatus, and more particularly to an image filing apparatus
providing image data suitable for output types of several
input/output devices.
2. Description of the Related Art
Generally, a scanner, a printer and a CRT are used mainly as
input/output devices in such an image filing apparatus.
There are many types of methods of displaying a tone image in use
for these input/output devices. For example, the binary method and
the multi value method are applied widely for the scanner. In the
printer, the binary method is used mainly, especially for the laser
printer, the binary method and the area tone method are applied
widely. In addition, there are many methods such as the binary
method, the multi value method and the color method for the CRT.
Thus, many types of methods of displaying a tone image are
available for input/output devices.
Conventionally, image data processed by such a method (the binary
method, the multi value method or the color method) are compressed
and extended using a standard data compression and extension
method. For example, the method (MH, MR, or MMR) using the Huffman
coding is well known for the compressing and extending process of
binary image data. Recently, a method referred to as arithmetic
coding has been used widely. In the compressing and extending
process multi-valued image data or color image data, generally,
two-dimensional orthogonal transforming method called as Adaptive
Discrete Cosine Transform (ADCT) is used widely. Thus, there are
several data compression and extension methods according to types
of input/output devices.
Usually, an image forming apparatus is provided with a plurality of
input/output devices, For example, a printer, a scanner, a CRT and
the like. As mentioned above, the binary method is most widely used
for a printer. Therefore, a transformation from a multi value to a
binary value is performed on data input through the scanner in an
image processing unit of the apparatus. Further, the display using
binary data must be used for the CRT in accordance with the
outputting type of the printer. The result is that there is a
decrease of the CRT's power of expression caused by the decrease of
the amount of information.
SUMMARY OF THE INVENTION
It is therefore an object of the present invention to provide an
image filing apparatus including a plurality of processing units
corresponding to types of compressing and extending methods,
respectively, in which one optimum processing unit is selected
according to the type of an assigned I/O device thereby to output
image data in the output form of the assigned I/O device such as an
image display, an image printer or the like.
The object of the invention, as will appear from a reading of the
following specification, is achieved by an image filing apparatus
comprising a unit for inputting an image, a plurality of data
compressing and extending units each for compressing the input
image data to store the compressed image data and each for
extending the compressed image data to output the extended image
data to at least one of image outputting units such as an image
display, an image printer and the like, and a control unit for
selecting one of the plurality of data compressing and extending
units in accordance with an image type of the input image to supply
it to the selected data compressing and extending unit upon the
compression and for supplying the compressed image data to the
selected data compressing and extending unit upon the
extension.
In operation, when compressing the input image data, the control
unit selects one of the plurality of data compressing and extending
units in accordance with an image type of the input image.
Preferably, the control unit comprises a unit for determining an
image type of the input image using the input image data and an
array of DCT (diacrete cosine transform) coefficients. Thus, the
optimum compressing and extending method is selected and applied to
the input image data in accordance with the type of the image
outputting unit from which the image data is to be output.
The object of the invention is also achieved by an image filing
apparatus comprising a unit for inputting an image, a data
compressing unit for compressing the input image to store the
compressed image data, a unit for extending the compressed image
data to output the extended data to at least one of image
outputting units such as an image display, an image printer and the
like, and a unit for determining an image type of an area of the
extended image data using the extended image data and DCT
coefficients in the area thereby to supply the extended image data
in the area, on which a predetermined image processing is performed
according to the image type, to at least one of image outputting
units.
When extending the compressed image data, that is the coded and
compressed image data is reconstructed, the determining unit
determines an image type of the compressed image data in areas
using the extended and decoded image data and DCT coefficients in
areas. Preferably, a plurality of predetermined image processings
are performed on the decoded image data in an area according to the
image type and one of them is selected and output to one of the
image outputting units.
Further, the object of the invention is attained by providing an
image filing apparatus comprising an unit for inputting an image, a
plurality of data compressing and extending units for compressing
the input image data to store a plurality of compressed data and
for extending the plurality of compressed data to output the
extended data to at least one of outputting units such as an image
display, an image printer and the like, respectively, and a control
unit for selecting one of the plurality of compressed data in
accordance with a type of the image outputting unit from which the
compressed image is to be output to supply the selected data to the
corresponding data compressing and extending unit upon the
extension and thereby to output the extended data to the image
outputting unit.
Further objects and advantages of the present invention will be
apparent from the following description of the preferred
embodiments of the invention as illustrated in the accompanying
drawings,
BRIEF DESCRIPTION OF THE DRAWINGS
FIG. 1 shows a block diagram of an embodiment of an image filing
apparatus according to the present invention;
FIG. 2 shows a flowchart diagram representing a flow of a main
process upon the compression in the image filing device according
to the invention;
FIG. 3 shows a flowchart diagram representing a flow of a main
process upon the extending operation for display in the filing
apparatus according to the invention;
FIG. 4 shows a flowchart diagram representing a flow of a main
process upon the extending operation for printing according to the
invention;
FIG. 5 shows a flowchart diagram representing a flow of a main
process upon the data compressing operations having a process of
changing the scale in the filing apparatus according to the second
embodiment;
FIG. 6 is a partially block diagram of the fourth embodiment of the
image filing apparatus according to the invention;
FIG. 7 is a partially block diagram of an example of the encoding
unit and the decoding unit of the image filing apparatus according
to the invention;
FIG. 8 shows pixel block in a case of dividing image information
into a plurality of 8.times.8 pixel blocks;
FIG. 9 shows an example of a quantization matrix;
FIG. 10 is a block diagram of the image processing unit of the
image filing apparatus according to the invention;
FIG. 11 is a timechart diagram of an example of the conventional
judging method:
Fig, 12 shows an example of DCT coefficients:
FIG. 13a shows mean magnitudes as for an image including an
intermediate tone area;
FIG. 13b shows mean magnitudes as for an image including
characters;
Fig, 13c shows mean magnitudes as for the image including meshes of
100 lines which is read by a scanner with 400 dpi;
FIG. 13d shows mean magnitudes as for the image including meshes of
200 lines which is read by a scanner with 400 dpi;
FIG. 14 shows the relationship between the target block for the
judgment and reference blocks:
FIG. 15 shows an example of divided areas of DCT coefficients in
use for the judgment of an image area with meshes:
FIG. 16 shows a partially block diagram of an example of the
judging unit 54 for an image area with meshes:
FIG. 17 shows an example of a scanning method in use for the
judgment of a peak located in the oblique direction of a matrix in
the judging unit 54;
FIG. 18a shows the mean magnitudes arranged in the third line of
the u direction of FIG. 13c;
FIG. 18b shows the mean magnitudes arranged in the third line of
the v direction;
FIG. 19 shows another example of a scanning method in use for the
judgment of a peak located in the oblique direction of the matrix
in the judging unit 54; and
FIG. 20 shows mean magnitudes of moduli of the AC components in the
oblique direction of an array of DCT coefficients.
DESCRIPTION OF THE PREFERRED EMBODIMENTS
FIG. 1 shows a block diagram of an embodiment of an image filing
apparatus according to the present invention. In FIG. 1, reference
numeral 1 denotes an optical reading device; 2, a printing device;
3, an image memory; 4, a CPU; 5, a keyboard; 6, a memory for
display; 7, a CRT; 8, a first compressing and extending unit; 9, a
second compressing and extending unit; 10, an image processing
unit; 11, an optical disk I/F circuit; and 12, an optical disk
drive device.
The filing apparatus according to the first embodiment is provided
with a plurality of compressing and extending units, that is the
first and second compressing and extending units 8 and 9 which
constitute the image data compressing and extending device of the
apparatus as shown in FIG. 1. Additionally, the apparatus is
characterized in that the CPU 4 is adapted to control the
compressing operation of image data and the extending operation of
the compressed data for a display or print operation.
The other constructions and operations are similar to those of a
conventional filing apparatus. The optical reading device 1 (for
example, a scanner) has a function for reading an original document
having characters, drawings or the like. The image memory 3 is a
memory for storing image data read by the reading device 1. That
is, the memory functions as an image storage unit.
The first compressing and extending unit 8 functions as one image
data compressing and extending unit for compressing and extending
image data using one compressing and extending method which is
suitable for the CRT 7, for example the DCT method.
The second compressing and extending unit 9 functions as the other
image data compressing and extending unit for compressing and
extending image data using another compressing and extending method
which is suitable for the printing device 2, for example the
Huffman coding method.
Further, the CRT 7 and the memory for display (for example VRAM)
are integral parts of an image display unit for outputting image
data and the printing device 2 is an image printing unit for
outputting image data.
The CPU 4 has a function for controlling the whole image filing
apparatus. The image processing unit 10 has a function for
processing image information in the apparatus. The optical disk
drive device 12 is a processing device comprising an optical disk
for storing document files and a driving unit for driving the
optical disk.
In this embodiment, the compressing operation using the plural
methods is performed (see FIG. 2) and the extending operation is
performed on the compressed image data with these methods (see
FIGS. 3 and 4).
Next, the description will be directed to these compressing and
extending operations referring to FIGS. 2 to 4.
FIG. 2 is a flowchart diagram representing a flow of a main process
upon the compression in the image filing device according to the
invention. In the figure, symbols #1-#3 show steps of the flow.
In the step #1, image data, which are read by the optical reading
device 1, are stored in the memory 3.
In the next step #2, the stored image data are coded by the first
compressing and extending unit 8 with one method which is suitable
for an image display unit such the CRT 7, for example the DCT
method, and then the coded data, which will be referred as first
coded data, are stored in the storing unit such as an optical
disk.
In the step #3, the stored image data are coded by the second
compressing and extending unit 9 with one method which is suitable
for the printing device 8, generally the Huffman coding method, and
then the coded data, which will be referred as second coded data,
are stored in the storing unit such as an optical disk.
In the aforementioned steps #1-#3, both the former compressed data,
or the first coded data available for an image display unit such
the CRT 7 and the latter compressed data, or the second coded data
available for the printing device 2 are stored in the optical disk
of the optical disk drive device 12.
Next, the description will be directed to the extending operation
for display.
FIG. 3 is a flowchart diagram representing a flow of a main process
upon the extending operation for display in the filing apparatus
according to the invention. In the figure, symbols #11-#13 show
steps of the flow.
In the step #11, the first coded data are read out of the storing
region of the optical disk or the like. The coded data are decoded
by the first compressing and extending unit 8 in the step #12. In
the next step #13, the decoded data are transferred to the memory
for display 6 and thereby displayed by the display unit such as the
CRT 7 or the like.
In the aforementioned steps #11-#13, the image data, which are
compressed with one method suitable for a display unit like the CRT
7, for example the DCT method, are extended (or decoded) to derive
data for display.
Thus, the image display unit in the form of a multi value system
allows the displacement of a tone image and provides a display easy
to see. In FIG. 1, the first compressing and extending unit 8 is
adapted for an image display unit like the CRT 7.
The description will be directed to the extending operation for
printing.
FIG. 4 shows a flowchart diagram representing a flow of a main
process upon the extending operation for printing according to the
invention. In the figure, symbols #21-#23 denote steps of the
flow.
In the step #21, the second coded data are read out of the storing
region of the optical disk or the like. The data are decoded by the
second compressing and extending unit 9 in the step #22. In the
next step #23, the decoded data are transferred to the memory (RAM)
and then are printed by the printing device 2.
In the aforementioned processing of the steps #21-#23, the image
data, which are compressed with one method suitable for the
printing device 2, are extended (or decoded) to derive data for
printing. In FIG. 1, the second compressing and extending unit 9 is
adapted for a printing unit like the printing device 2.
Generally, the Huffman coding method is available for the printing
device 2 because the binary method is widely used.
If the printing device 2 is in the form of the area tone method,
the first compressing and extending unit 8 can be used upon the
compressing and extending operations for the device 2.
In an example of a filing apparatus for moving pictures provided
with a plurality of image processing units and a plurality of
compressing and extending units corresponding to the image
processing units, the similar constructions are provided.
Next, the description will be directed second embodiment of the
filing apparatus according to the invention.
In this embodiment, the construction of the filing apparatus is
similar to that of the first embodiment shown in FIG. 1. The
extending operations for display and printing are similar to those
of the first embodiment aforedescribed in FIGS. 3 and 4, but only
the operations for data compressing are different from those of the
first embodiment.
FIG. 5 shows a flowchart diagram representing a flow of a main
process upon the data compressing operations having a process of
changing the scale of an image in the filing apparatus according to
the second embodiment. In the figure, symbols #31-#34 denote steps
of the flow.
In the step #31, image data, which are read out of the optical
reading device 1, are stored in the memory (RAM). The stored data
will be referred as first image data. In the step #32, the scale of
the first image data are modified by the image processing unit 10
in accordance with the number of pixels of the display unit,
thereby the second image data are derived. In the step #33, the
second image data are encoded by the first compressing and
extending unit 8 and the coded data are stored as first coded data
in the storing region of the optical disk or the like. In the next
step #34, the first image data are encoded by the second
compressing and extending unit 9 and the coded data are stored as
second coded data in the storing region of the optical disk or the
like.
In the aforementioned processing of the steps #31-#33, the scaling
operation is performed in accordance with the number of pixels of
the image display unit.
Next, the description will be directed to a third embodiment of the
filing apparatus according to the invention.
In this embodiment, the basic construction is similar to that of
the first embodiment shown in FIG. 1.
The apparatus according to the third embodiment is provided with a
storing region in a RAM for storing an instruction input by an user
with the keyboard 5 or the like. The instruction indicates one of
compressing and extending methods available in the apparatus. That
is, the keyboard 5 functions as a selecting means for selecting one
of plural compressing and extending methods. The instruction input
with the keyboard is stored in the RAM, and then the compressing
and extending processes follow the instruction. Thus, this
embodiment can allow an user to select a desirable compressing and
extending method.
Next, the description will be directed to a fourth embodiment of
the image filing apparatus according to the invention.
This embodiment is characterized in that there is provided with
means for discriminating between types of original documents, that
is means for determining types of original documents. When
compressing image data, one of plural compressing methods is
selected automatically using the discriminating means.
FIG. 8 is a partially block diagram of the fourth embodiment of the
image filing apparatus according to the invention. In this figure,
the numeral number 21 denotes an image reading unit; 22, an
encoding unit; 23, a decoding unit; 24, an image processing unit;
25, an image outputting unit; and 26, a code storage unit.
Original documents are read by the image reading unit 21 having a
CCD (charge-coupled device) or the like. The read analog image data
are encoded by the encoding unit 22 using the ADCT method. The
detail construction of the encoding unit 22 will be described
later. The encoded image data are stored in the code storage unit
26 which is constituted by a storage device such as a magnetic disk
device or the like.
When reconstructing the corresponding images from the encoded data
of the original documents, the encoded data are decoded by the
decoding unit 23 to derive a signal of the reproduced image and the
signal is output to the image processing unit 24.
In the image processing unit 24, the input signal of the
reconstructed data is processed using an intermediate tone process
and/or a simple binary process which may be selected before the
processing. The processed data are output to the image outputting
unit 25 such as a binary printer, a CRT or the like, for producing
a hard copy and displaying the data on a display.
In this case, the image processing unit 24 uses magnitudes of
decoded image data which are produced within the decoding process
in the decoded unit 23 for the purpose of judging whether the
encoded image has an intermediate tone image area.
Thus, the image filing apparatus determines whether encoded image
data include an intermediate tone. Preferably, the discrimination
is performed upon the filing and the results of the discrimination
are stored as one file searching condition in a certain area of the
optical disk. When searching a file, the corresponding result of
the discrimination are displayed as one file searching condition on
the display.
Next, the description will be directed to the detail construction
and the operation of the encoding unit 22 and the decoding unit 23
shown in FIG. 6.
FIG. 7 is a partially block diagram of an example of the encoding
unit and the decoding unit of the image filing apparatus according
to the invention. In the figure, the numeral number 31 denotes a
block reading unit; 32, a DCT transforming unit; 33, a quantizing
unit; 34, a Huffman coding unit; 35, a scale factor control unit;
36, a first calculating unit; 37, a quantization matrix storage
unit; 38, a second calculating unit; 39, a code storage unit; 40, a
Huffman decoding unit; 41, a inverse quantizing unit; 42, a inverse
DCT transforming unit; and 43, a block writing unit.
The description will be directed to the Discrete Cosine Transform
(so-called DCT) coding.
In the coding process, image data are read out every N.times.N
pixel block (for example, 8.times.8 pixel block) by the block
reading unit 31, and then those block data are converted into DCT
coefficients, respectively.
Now, an example of processing image data in 8.times.8 pixel blocks
will be described.
FIG. 8 shows pixel blocks in the case of dividing image information
into a plurality of 8.times.8 pixel blocks.
The image information is transformed into DCT coefficients every
block as shown in FIG. 8. This DCT transformation is given by the
following expression. ##EQU1## for u, v=0, 1, . . . , 7, where
x.sub.ij denotes an original pixel's data; Y.sub.uv' a coefficient
for the discrete cosine transform; N=8; and ##EQU2##
The coefficient matrix are output from the DCT converting unit 32
to the quantizing unit 33. The coefficient matrix is linearized
with quantization step sizes each having a different amount for
each of the coefficients, and thereby quantization coefficients are
derived. The first calculating unit 36 multiplies a quatization
matrix stored in the quantization matrix storage unit 37 by a scale
factor preset in the scale factor control unit 35 to derive those
quantization step sizes corresponding to the coefficients,
respectively.
FIG. 9 shows an example of a quantization matrix. The horizontal
axis represents u in the aforedescribed equation and the vertical
axis represents v in the figure. The quantization matrix is adapted
to quantize coefficients in a lower order more quatization levels
and code bits in accordance with the visual characteristics of
human beings. A quantity of coding or a quality of a decoded image
can be controlled by varying a value of the aforementioned scale
factor.
Among the quantized coefficients, which are quantized by the
quantizing unit 33, some of the coefficients with a higher
frequency are allocated less code bits than the other coefficients
with a lower frequency, thereby decreasing the total amount of
coding. The quantized coefficients are stored in the code storage
unit 39.
When outputting an image, that is, performing a decoding process,
codes are read out of the code storage unit 39 and the read codes
are delivered to the Huffman decoding unit 40. Each value
corresponding to each of the quantized coefficients is assigned to
each code in the Huffman decoding unit 40 and the value is
inverse-quantized in the inverse-quantizing unit 41.
In this case, the second operation unit 38 computes a step size for
quantization in this inverse-quatization by multiplying the
quatization matrix stored in the quatization matrix storage unit 37
by the scale factor preset by the scale factor control unit 35.
The DCT coefficients output by the inverse quantization unit 41 are
transformed to the corresponding image data in the inverse-DCT
transforming unit 42 and then the image data are output every block
by the block writing unit 43.
The image data, which are output in blocks, are stored temporally
in one block line memory. The one block line memory can store a row
of pixel blocks in the scanning direction. In this embodiment, the
one block line memory is composed of 8 line memory areas
corresponding to 8 lines in an image because one pixel block
includes 8.times.8 pixels. The image data are output every one line
to the image processing unit 24 in FIG. 6 when the one block line
memory is fulfilled with one line of 8.times.8 pixel blocks.
The image processing unit performs a simple binary processing or an
intermediate tone processing on the output image data in parallel
to output the processed data to the image output unit 25.
Next, the description will be directed to a detail of the image
processing unit 24 in FIG. 6.
FIG. 10 is a block diagram of the image processing unit of the
image filing apparatus according to the invention. In this figure,
the reference numeral 51 denotes an intermediate tone processing
unit; 52, a simple binary processing unit; 53, a judging unit for
an intermediate tone image area; 54, a judging unit for an image
area with meshes; 55, an OR gate circuit; 56, an output control
unit; and 57, an image output unit.
As shown in FIG. 10, the image processing unit reads decoded image
data and performs an intermediate tone process on the data in the
intermediate tone processing unit thereof using the DTER method,
the density pattern method, the submatrix method or the like and in
parallel performs a simple binary process on the data in the binary
processing unit 52 thereof.
The output control unit 56 selects either a signal from the
intermediate tone processing unit 51 or a signal from the simple
binary processing unit 52 in each area of the image by means of the
output signal from the OR gate circuit 55 and outputs the selected
one to the image output unit 57. The intermediate tone area judging
unit 53 judges whether decoded image data in question have an
intermediate tone area using the decoded image data and outputs a
signal with "1" state when the data includes an intermediate tone
area. The judging unit for an image area with meshes 54 judges
whether the decoded image data have an meshes area and outputs a
signal with "1" state when the judgment result is "yes". These
signals are input to the OR gate circuit 55.
The output control unit 56 selects the output signal from the
intermediate tone processing unit 51 when the output signal of the
OR gate is "1" state and selects the output signal from the simple
binary processing unit 52 except that.
A judging method of judging whether image data have an intermediate
tone area is known (for example, Japanese Laid Open No. 61-146067
(KOKAI)). In this method, an AND operation is performed on an
original signal representing decoded image data and the shifted
signal delayed by plural pixels in the main scanning direction.
Next, the description will be directed to the conventional judging
method.
FIG. 11 is a timechart diagram of an example of the conventional
judging method. In this figure, th1 on the vertical axis represents
a threshold value for the simple binary processing; th2, a binary
threshold value for the intermediate tone judging process. The
symbols B1 represents an area with light and shade tone; B2, a
thick line; and B3, a thin line. The symbols a-g show signals.
For example, information with respect to a digital image is
obtained along the main scanning direction at a sub scanning
position as shown in an upper part of FIG. 11. The digital image
information is binarized with the threshold value th2 thereby to
produce the signal shown with the symbol a in FIG. 11. The signal a
is delayed by 8 pixels (8p) to derive the signal b. The AND
operation is performed on the signals a and b thereby to produce
the signal shown with the symbol c. This signal c is utilized for
the judgment of the intermediate tone.
On the other hand, the simple binary processing unit 52 generates
the signal d in FIG. 11 with the threshold value th1 for the binary
processing. This signal d is delayed by 4 pixels (4 p) to derive
the signal e. The timing of the signal e matches with the signal c
for the intermediate tone judging. In addition, the intermediate
tone processing unit 51 outputs the signal f delayed by 4 pixels (4
p) from the signal a,
Meanwhile, it is difficult to illustrate a signal representing the
intermediate tone, however, the signal f is provided in FIG. 11 to
explain the intermediate tone processing in general. Either the
signal e or the signal f is selected by the output control unit 56
with the signal c for the intermediate tone judgment to obtain the
signal g in the FIG. 11.
In the signal g, the hatching parts represent the output signal by
the simple binarizing process and the other parts represent the
output signal by the intermediate tone process. The light and shade
tone part B1 in the left side of FIG. 11 corresponds to a signal
processed with the intermediate method and the thin line part B2
corresponds to a signal processed with the simple binarizing
method.
On the other hand, the both side parts of the thick line part B2,
which are in the central part of FIG. 11, correspond to a signal
processed with the simple binarizing method and the other parts
thereof correspond to a signal processed with the intermediate tone
method.
Therefore, the conventional intermediate tone judging system can
discriminate between characters and intermediate tone areas which
are mixed in one image thereby to perform a preferable processing
on each of them.
However, when an output of the output control unit 56 is selected,
a signal output from the simple binarizing unit 52 is selected for
an image area including meshes in the conventional system.
Such a disadvantage is dissolved by this embodiment in which the
judging unit 54 for an image area with meshes and the OR gate
circuit 55 are disposed. The judging unit 54 outputs a signal
having "1" state in an area with meshes and having "0" state in
another area. The signal are input to the OR gate circuit 55
together with the output signal from the intermediate tone judging
unit 53. The result is that a signal output from the intermediate
tone processing unit 51 can be selected for areas having
meshes.
Next, the detail description will be directed to the judging unit
54 shown in FIG. 7.
The judging unit 84 for an image area with meshes receives DCT
coefficients output from the inverse quantizing unit 41 and judges
whether an area of an image in question includes meshes on the
basis of the DCT coefficients,
FIG. 12 shows an example of DCT coefficients. The coefficient (0,0)
in the upper left-hand corner of each block represents a mean
density of the pixel block and coefficients in the lower order
located adjacent to the coefficients (1, 0) and (0, 1) represent a
rate of lower frequency components involved in the pixel block as
shown in FIG. 12. On the other hand, coefficients in the higher
order represent a rate of higher frequency components in the pixel
block.
In view of the aforementioned relationship, the coefficient (0, 0)
is called the DC component and the others are called the AC
components. In addition, the coefficients (1, 0), (2, 0), (3, 0), .
. . in the u direction (main scanning direction) represent a
variation of densities in the main scanning direction. Similarly,
the coefficients (0,1), (0,2), (0,3), . . . in the v direction (sub
scanning direction) represent a variation of densities in the sub
scanning direction. Further, the coefficients (1, 1), (2, 2), (3,
3), . . . in the oblique direction represent the magnitudes of the
components where the variation of the density in the main scanning
direction and the variation of the density in the sub scanning
direction are superimposed.
Now, the description will be directed to an example of DCT
coefficients. When the cosine transform operation is performed on
pixel blocks as for an image including intermediate tone areas or
characters to derive mean magnitudes of moduli of the AC
components, respectively, the magnitudes of them are illustrated in
FIG. 13.
FIG. 13a shows mean magnitudes as for an image including
intermediate tone areas, FIG. 13b shows mean values as for an image
including characters, FIG. 13c shows mean magnitudes as for an
image including meshes of 100 lines which is read by a scanner with
400 dpi, and FIG. 13d shows mean values as for an image including
meshes of 200 lines which is read by a scanner with 400 dpi.
As seen clearly from the comparison between FIG. 13a and FIG. 13b,
they are different from each other with respect to the magnitudes
of the higher frequency components, that is, the magnitudes of the
character image are larger than those of the intermediate image,
but both the magnitudes decrease gradually at a higher frequency.
On the other hand, the magnitudes of DCT coefficients as for an
image including meshes, which are obtained with the similar means,
show specific patterns in which the magnitudes vary along both the
main scanning and the sub scanning directions. For example, the
peak of the magnitudes of AC components, which is emerged in the
oblique direction of the matrix, is located in a position along the
direction dependent on the frequency of meshes as shown in FIGS.
13c and 13d.
The filing apparatus according to the embodiment judges whether an
area of an image includes meshes using this specific characteristic
as for an image including meshes.
FIG. 14 shows relationship between a target block for the judgment
and reference blocks. When judging whether the target block for the
judgment is an image area including meshes, 5 blocks in the
preceding line, 2 preceding blocks in the same line as the target
block and 2 next blocks in the same line, that is a total of 9
reference blocks are also used. The mean magnitudes of the moduli
of the AC components are obtained from those DCT coefficients in
the target block and the nine reference blocks, respectively, and
then it is judged whether there is a peak, which is located in the
oblique direction of the matrix of the mean magnitudes,
characteristic of an image with meshes.
FIG. 15 shows an example of divided areas of DCT coefficients in
use for the judgment of an image area with meshes. In the figure,
the symbols 1-5 denote the areas of DCT coefficients.
The five areas 1-5 are preset as shown in FIG. 15.
The coefficients (2, 2), (3, 2), (2, 3) and (3, 3) are involved in
the first area 1. The coefficients (3, 3), (4, 3), (3, 4) and (4,
4) are involved in the second area 2. The coefficients (4, 4), (5,
4), (4, 5) and (5, 5) are involved in the third area 3. The
coefficients (5, 5), (6, 5), (5, 6) and (6, 6) are involved in the
fourth area 4. The coefficients (6, 6), (7, 6), (6, 7) and (7, 7)
are involved in the fifth area 5.
As for each of the first to fifth areas 1-5, it is judged whether a
peak emerges or not. That is, by comparison between the mean
magnitude of coefficients in one of the areas 1-5 and those in the
preceding coefficients each having a lower frequency, it is judged
whether said one area has a peak or not.
Concretely, the mean magnitude of the area 1 is compared with those
of the coefficients (1, 1), (2, 1), (3, 1), (1, 2) and (1, 3). The
mean magnitude of the area 2 is compared with those of the
coefficients (2, 2), (3, 2), (4, 2), (2, 3) and (2, 4). The mean
magnitude of the area 3 is compared with those of the coefficients
(3,3), (4,3), (5,3), (3,4) and (3,5). The mean magnitude of the
area 4 is compared with those of the coefficients (4, 4), (5, 4),
(6, 4), (4, 5) and (4, 6). The mean magnitude of the area 5 is
compared with those of the coefficients (5, 5), (6, 5), (7, 5), (5,
6) and (5, 7) . Then, as for one of the areas, if the corresponding
mean magnitudes of the preceding coefficients are smaller than
those of the area in question, it is concluded that the target
block includes a meshes image.
FIG. 16 shows a partially block diagram of an example of the
judging unit 54 for an image area with meshes. In the figure, the
reference numeral 61 denotes a two-blocks line memory; 62, a mean
magnitude processing unit; 63, a judging unit; and 64, an one-block
line memory.
In FIG. 9, the DCT coefficient data, which are inverse quantized
and reproduced, are stored temporally in the two-blocks line memory
61. Then, the DCT coefficients of one pixel block which is a target
block for the judgment and the DCT coefficients of the other
reference blocks in the relationship with the target block as shown
in FIG. 14 are read out and output serially to the mean magnitude
processing unit 62.
The mean magnitude processing unit 62 outputs the 8.times.8 pixel
block each element of which is a mean magnitude of the moduli of
the corresponding coefficients of the reference blocks and the
target block for the judgment. The judging unit 63 judges whether
any one of the areas 1-5 in the 8.times.8 pixel block includes the
peak or not from the aforementioned operation. When it is concluded
from the judgment that one of the areas includes the peak, the
judging unit 63 outputs the signal with "1" state for the 8.times.8
pixel block, whereas the judging unit 63 outputs the signal with
"0" state for the 8.times.8 pixel block when any one of the areas
includes no peak.
The output signal with "1" state or "0" state is stored in the next
one-block line memory 84. The one-block line memory 64 of the
judging unit 54 for an image area with meshes outputs the signal
with "1" state or "0" state every one line in synchronism with the
signal output by the judging unit 53 for an intermediate tone image
area shown in FIG. 10. Then, those signals are input to the OR gate
circuit 55 and processed therein to supply the processed signal to
the output control unit 56. Therefore, when an area of an image
including meshes or intermediate tone is processed, the processed
signal with the intermediate tone method is selected by the output
control unit 56 and output to the image output unit. Thus, the
judging signal for an image area, which indicates that an image
area includes meshes and represents one characteristic of image
information, is obtained.
Another example of the judging method, by which the meshes area
judging unit 54 judges whether an area of an image includes a peak
of DCT coefficients characteristic of meshes and located in the
oblique direction of the matrix, will be described hereinafter.
FIG. 17 shows an example of a scanning method in use for the
judgment of a peak located in the oblique direction of a matrix in
the judging unit 54 for an image area with meshes. In the figure,
the symbols 3-8 denote column's number and row's number.
As for mean magnitudes of moduli of the AC components of DCT
coefficients in a target block for the judgment, scanning
horizontally and vertically the mean magnitudes in the 8.times.8
pixel block from the third column and the third row as shown in
FIG. 17 searches an element which satisfies such a condition that
the element is located in the oblique direction of the matrix, and
the magnitude is larger than 1.5 times the magnitude of the
preceding element at the same column and than 1.5 times the
magnitude of the preceding element at the same row. If such an
element is present, it is concluded that there is a peak in the
oblique direction of the matrix.
FIG. 18a shows the mean magnitudes arranged in the third line of
the u direction of FIG. 13c and FIG. 18b shows the mean magnitudes
arranged in the third line of the v direction.
For example, when the judgment is performed on the case of FIG. 13c
using this method, it is clear that there is a peak at the third
row and the third column as seen from FIG. 18 in which the
magnitudes are arranged in the third line of the u direction and
arranged in the third line of the v direction. The magnitude in the
third element in the third line of the u direction is larger than
that of the preceding element in the same line and the magnitude in
the third element in the third line of the v direction is larger
than that of the preceding element in the same line.
Therefore, it is concluded that the image area having such a peak
includes meshes.
In a preferred embodiment, another judging method is available.
FIG. 19 shows another example of a scanning method in use for the
judgment of a peak located in the oblique direction of the matrix
in the judging unit 54 for an image area with meshes.
As for mean magnitudes of moduli of the AC components of DCT
coefficients in a target block for the judgment, scanning obliquely
the mean magnitudes in the 8.times.8 pixel block from the third
column and the third row as shown in FIG. 19 searches an element
which satisfies such a condition that the magnitude is larger than
1.5 times the magnitude of the preceding element. If such an
element is present, it is concluded that there is a peak at a
diagonal element of the matrix.
FIG. 20 shows mean magnitudes of moduli of the AC components in the
oblique direction of an array of DCT coefficients. In the figure,
the symbols 1-8 denote the line number.
For example, when the judgment is performed on the case of FIG. 13c
using this method, it is clear that there is a peak at the third
row and the third column as seen from FIG. 20 in which the mean
magnitudes are arranged in the oblique direction. The magnitude in
the third diagonal element is larger than that of the preceding
element. Therefore, it is concluded that the image area having such
a peak includes meshes.
Thus, the judgment of areas having meshes is performed using the
aforementioned judging method thereby to generate a judging signal
representing whether an area in question includes meshes or
not.
Since the apparatus according to the invention is provided with a
means for determining a type of originals aforedescribed in detail
with FIGS. 6 to 20, a compressing method suitable to the type of
originals can be selected automatically through the judgment result
by the determining means when compressing the originals.
Concretely, an area for storing the judgment result is provided and
the data representing the judgment result is referred, and thereby
the desired compressing method is selected.
Many widely embodiments of the present invention may be constructed
without departing from the spirit and the scope of the present
invention is not limited to the specific embodiments described in
the specification, except as defined in the appended claims.
* * * * *